2026-08论文中文摘要
MNC：面向私有大语言模型智能体通信的范围受限语义解密
MNC: Scope-Bound Semantic Declassification for Private LLM-Agent Communication
arXiv 2608.01719 · 2026-08-04LaCache：面向大语言模型服务的鲁棒语义缓存
LaCache: Robust Semantic Caching for LLM Serving
arXiv 2608.01718 · 2026-08-04带符号图的Lovász Theta参数与Theta体
A Lovász Theta Parameter and Theta Body for Signed Graphs
arXiv 2608.01716 · 2026-08-04编码智能体作为测试套件审计器：在接近官方测试套件检测能力的同时发现官方套件遗漏的问题
Coding Agents as Test-Suite Auditors: Finding What Official Suites Miss While Approaching What They Catch
arXiv 2608.01715 · 2026-08-04STC-Net：基于电致发光的太阳能电池裂纹分割用于功率损失估计
STC-Net: Electroluminescence-Based Solar Cell Crack Segmentation for Power Loss Estimation
arXiv 2608.01714 · 2026-08-04广义信道状态信息反馈的时域信道估计
Temporal Channel Estimation for Generalized CSI Feedback
arXiv 2608.01713 · 2026-08-04气相中碰撞后热激发与受存活限制的团簇生长
Post-Collision Thermal Excitation and Survival-Limited Cluster Growth in the Gas Phase
arXiv 2608.01712 · 2026-08-04使用智能体管控系统构建用于元分析综合的可执行分析知识表示
Constructing Executable Analytical Knowledge Representations for Meta-Analysis Synthesis Using an Agentic Harness
arXiv 2608.01711 · 2026-08-04超越单次使用令牌：面向抗重放的大语言模型智能体行动的持久授权状态
Beyond Single-Use Tokens: Durable Authorization State for Replay-Resistant LLM Agent Actions
arXiv 2608.01710 · 2026-08-04SpatialQuery：评估视觉语言模型中基于几何的多实例空间推理能力
SpatialQuery: Benchmarking Geometry-Grounded Multi-Instance Spatial Reasoning in Vision-Language Models
arXiv 2608.01709 · 2026-08-04拓扑在机器学习训练性能中的作用
On Topology's Role in ML Training Performance
arXiv 2608.01707 · 2026-08-04UniSim-SLAM：采用统一Sim(3)优化的前馈SLAM
UniSim-SLAM: Feed-Forward SLAM with Unified Sim(3) Optimization
arXiv 2608.01706 · 2026-08-04重新思考生成式AI素养：面向K-12教师教育的整合型、发展型与辩证型框架
Rethinking Generative AI Literacy: An Integrative, Developmental, and Dialectical Framework for K-12 Teacher Education
arXiv 2608.01705 · 2026-08-04下限、上限与融合差距：机器能预测多少群体阅读注意力？
Floor, Ceiling, and the Fusion Gap: How Much of Crowd Reading Attention Can Machines Predict?
arXiv 2608.01704 · 2026-08-04具有自适应攻击者序列与时变攻击强度的Stackelberg安全博弈在线学习
Online Learning in Stackelberg Security Games with Time-Varying Attack Intensities
arXiv 2608.01703 · 2026-08-04带夸克的线性σ模型中磁场对π介子-夸克顶点的修正
Magnetic field modifications to the pion-quark vertex in the Linear Sigma Model with quarks
arXiv 2608.01701 · 2026-08-04勒夫谢茨束的特征簇上的单值群作用
Monodromy action on character varieties for Lefschetz pencils
arXiv 2608.01700 · 2026-08-04论正非线性系统的输入输出持久性与互联
On input-output persistency and the interconnection of positive nonlinear systems
arXiv 2608.01699 · 2026-08-04表面分子体系中的针尖增强振动阶梯爬升
Tip-Enhanced Vibrational Ladder Climbing in Surface Molecular System
arXiv 2608.01698 · 2026-08-04面向以球员为中心的球动作检测的实体感知序列转换
Entity-Aware Sequence Transduction for Player-Centric Ball Action Spotting
arXiv 2608.01696 · 2026-08-04分层开放通量管中的阿尔文运动：从传播到局地驻波的转变及对开尔文-亥姆霍兹不稳定性（KHI）的启示
Alfvénic Motions in a Stratified Open Flux Tube: Transition from Propagating to Locally Standing Motions and Implications for the Kelvin-Helmholtz Instability
arXiv 2608.01695 · 2026-08-04ARM：具有有限样本误差控制的检测器无关变点归因方法
ARM: Detector-Agnostic Changepoint Attribution with Finite-Sample Error Control
arXiv 2608.01691 · 2026-08-04ProtoAct：将湿实验室协议转化为具身机器人动作
ProtoAct: Turning Wet-Lab Protocols into Embodied Robotic Actions
arXiv 2608.01690 · 2026-08-04Orlicz-BMO交换子在Orlicz-Hardy型空间上的端点有界性
Endpoint boundedness of Orlicz-BMO commutators on Orlicz-Hardy type spaces
arXiv 2608.01689 · 2026-08-04超越基于得分的游戏化：为VR颈部康复设计时空与音乐体验
Beyond Score-Based Gamification: Designing Spatiotemporal and Musical Experiences for VR Neck Rehabilitation
arXiv 2608.01688 · 2026-08-04仿射空间束在原点处的扎里斯基余切空间
The Zariski Cotangent Space at the Origin of the Pencil Diffeological Space
arXiv 2608.01687 · 2026-08-04医学影像中的生成式AI与基础模型
Generative AI and Foundation Models in Medical Image
arXiv 2608.01686 · 2026-08-04用于相关分类结局的带ANOVA分解的多变量多项Logit模型
Multivariate Multinomial Logit Model with ANOVA Decomposition for Correlated Categorical Outcomes
arXiv 2608.01685 · 2026-08-04GABench：用于评估大语言模型智能体图分析任务的综合基准
GABench: A Comprehensive Benchmark for Evaluating LLM Agents on Graph Analysis Tasks
arXiv 2608.01684 · 2026-08-04调和函数边界唯一延拓的光滑失效
Smooth Failure of Boundary Unique Continuation for Harmonic Functions
arXiv 2608.01683 · 2026-08-04有限反馈RIS辅助系统的无信道估计波束成形
Channel-Estimation-Free Beamforming for RIS-Assisted Systems with Limited Feedback
arXiv 2608.01681 · 2026-08-04通过反事实评估理解长上下文基础模型中的稀疏注意力选择性
Understanding Sparse Attention Selectivity in Long-Context Foundation Models via Counterfactual Evaluation
arXiv 2608.01676 · 2026-08-04具有属于稳定律吸引域的新息的回火线性过程的极限定理
Limit Theorems for Tempered Linear Processes with Innovations in the Domain of Attraction of a Stable Law
arXiv 2608.01674 · 2026-08-04学习该记住什么：基于上下文蒸馏的测试时训练
Learning What to Remember: Test-Time Training via Context Distillation
arXiv 2608.01672 · 2026-08-04Erdős问题725的次线性范围内拉丁方的Godsil–McKay渐近公式
The Godsil--McKay Asymptotic for Latin Rectangles in the Sublinear Range of Erdős Problem 725
arXiv 2608.01671 · 2026-08-04基于投影关联与有序根定律的尖锐根反集中性
Sharp Root Anti-Concentration via Projective Incidence and Ordered Root Laws
arXiv 2608.01670 · 2026-08-04面向5G非地面网络切片的低地球轨道（LEO）感知深度强化学习元调度器
LEO-Aware DRL Meta-Scheduler for 5G Non-Terrestrial Network Slicing
arXiv 2608.01668 · 2026-08-04TCPO：回合级信用策略优化
TCPO: Turn-Level Credit Policy Optimization
arXiv 2608.01667 · 2026-08-04排序前分配：面向通用大模型（OmniLLMs）的解耦令牌压缩
Allocation Before Ranking: Decoupled Token Compression for OmniLLMs
arXiv 2608.01665 · 2026-08-04FAU参加2026年ImageCLEF多模态推理任务：鲁棒候选评分与简洁多语种视觉问答
FAU at ImageCLEF 2026 Task on Multimodal Reasoning Robust Candidate Scoring and Concise Multilingual Visual Answering
arXiv 2608.01664 · 2026-08-04基于视觉定位的分割基础模型少样本概念提示学习
Few-Shot Concept Prompt Learning for Segmentation Foundation Models via Visual Grounding
arXiv 2608.01663 · 2026-08-04FairForensics：基于视觉-语言建模的通用公平深度伪造检测——表情感知与人口统计解析
FairForensics: Seeing Expressions and Parsing Demographics via Vision-Language Modeling for Generalizable Fair Deepfake Detection
arXiv 2608.01661 · 2026-08-04锚定、覆盖与优化：面向长视频问答的以证据为中心的帧选择框架
Ground, Cover, and Refine: Evidence-Centric Frame Selection for Long-Video Question Answering
arXiv 2608.01660 · 2026-08-04在多租户大语言模型前缀缓存中保留准入责任
Preserving Admission Responsibility in Multi-Tenant Large Language Model Prefix Caches
arXiv 2608.01657 · 2026-08-04你无法优化你无法衡量的：多任务评估作为AI介导的平视交互的缺失基础
You Cannot Optimize What You Cannot Measure: Multitasking Evaluation as the Missing Foundation of AI-Mediated Heads-Up Interaction
arXiv 2608.01656 · 2026-08-04PrefixPlace：异构计算与传输成本下大语言模型服务的可证明前缀键值放置方案
PrefixPlace: Provable Prefix Key-Value Placement for Large Language Model Serving under Heterogeneous Compute and Transfer Costs
arXiv 2608.01655 · 2026-08-04用于相位恢复的带随机初始化的近似消息传递
Approximate Message Passing with Random Initialization for Phase Retrieval
arXiv 2608.01654 · 2026-08-04用于学习型图像编码的哈达玛域模型量化
Hadamard-Domain Model Quantization for Learned Image Coding
arXiv 2608.01653 · 2026-08-04SyncPlan：通过显式同步与自适应修正实现长视 Large Language Model（大语言模型，LLM）协调
SyncPlan: Long-Horizon LLM Coordination with Explicit Synchronization and Adaptive Correction
arXiv 2608.01652 · 2026-08-04Bole：面向混合注意力语言模型的高效树推测解码
Bole: Efficient Tree Speculation for Hybrid-Attention Language Models
arXiv 2608.01651 · 2026-08-04
